Conclusions

Analyzing and simulating experiential capacity at a wild and scenic river

Geoffrey K. Riungu ^a*, Jeffrey C. Hallo^a, J. Adam Beeco^b, Lincoln R. Larson^c, Matthew T. J.

Brownlee^a, Kenneth F. Backman^a, Unmesh Kanchan^d, & Fefer, P. Jessica^a

a Clemson University, Department of Parks, Recreation and Tourism Management, Lehotsky Hall 142 Jersey Ln, Clemson, SC 29631

b National Park Service, National Resource Stewardship and Science Directorate, 1201 Oakridge Dr, Suite 100, Fort Collins, CO 80525

c North Carolina State University, Department of Parks, Recreation and Tourism Management, Raleigh, NC 27695-8004

d Clemson University, School of Computing, 821 McMillan Rd, Clemson, SC 29631

*Corresponding author

Email address: [email protected] (G.K. Riungu).

Abstract

Recreational boating is one of the major water-based activities in the U.S. Visitors enjoy a multitude of public navigable rivers for boating, but what is the capacity of these waterways to accommodate use? An agent-based simulation model of a popular Wild and Scenic River was developed to determine the river’s social carrying capacity. The findings show that the current boat use levels for an average non-peak were well within acceptable thresholds for boating. Also, at current boat use level for an average peak day, very few cases of boating thresholds being violated were recorded. However, increasing boat use levels by over 25% may result in

“acceptability” and “displacement” thresholds to be exceeded at certain portions of the day. The study identified that one of three river use zones may experience a larger proportion of

crowding-related threshold violations. Also, by applying perceived crowding thresholds for boating derived from a 800m viewshed across river segments that were not uniformly sized, the study examined the practicality of applying crowding thresholds for boating in terms of boat density or merely the number of boats that are within the line of sight of visitors at any given time. The results show that there were no significant differences in simulation outcomes for crowding thresholds for boating between the river segments (i.e., viewsheds that were 800m and smaller sized ones). Therefore, the maximum number of boats that are within the line of sight of visitors at any given time may be a more logical threshold to apply across an entire river because viewsheds are likely to change with the river’s morphology. The study demonstrated a proof of concept in simulating recreational boating and makes significant contribution to the body of knowledge by applying norm-based approaches to determine acceptable boating conditions.

Also, the study findings can inform the development of visitor use management plans for public

waterways and assist managers who seek to promote a high quality recreational boating experience.

Key words: Recreational boating, simulation, agent-based model, NetLogo, carrying capacity, thresholds

Visitation to protected areas (PAs) has steadily increased with countries like the U.S. National Park System recording visitation at over 330 million visits in 2017

(National Park Service [NPS], 2018). With a significant portion of navigable waterways located in PAs, recreational boating (RB) has also experienced a steady increase. For example, paddle sports such as stand-up paddling and whitewater kayaking reported increasing average participation by 26% and 10%, respectively, from 2012 to 2015 (Outdoor Foundation, 2016).

Increased boating levels raise safety issues, can create shoreline erosion, and strain infrastructure and facilities such as boat ramps and parking areas (Itami, Gimblett

& Poe, 2017). Therefore, determining acceptable levels of impact from visitor use (i.e., carrying capacity) is a perennial issue on waterways such as rivers due to: 1) increasing boating participation rates that may jeopardize the integrity of natural resources and the quality of the visitor experience (Manning, 2011), and 2) regulatory or planning

requirements such as the Wild and Scenic Rivers Act of 1968 that mandates federal agencies managing rivers designated as wild, scenic or recreational to address resource and experiential impacts associated with visitation, specifically including setting capacities. Additionally, enabling legislation such as the Organic Act of 1916 indicate that the NPS must provide for public opportunities to enjoy a park unit’s natural and cultural resources. This has spurred a need to consider the visitor experience, its management, and the related issue of carrying capacity as a core element of any park’s efforts.

To effectively manage PAs and plan for their future, a basic understanding of a park’s resources, values, uses, and users is needed. For the latter two, the development of a visitor use management plan provides detailed overall guidance on management

strategies, including capacities, associated with particular visitor use facilities, visitor activities, and visitor use issues. Planning frameworks such as the Visitor Use

Management (VUM) framework are designed and intended to help guide all visitor use management on federal agency lands and waters (Interagency Visitor Use Management Council [IVUMC], 2016). In the face of increasing visitor use, the VUM framework is aimed at maintaining the quality of the visitor experience while protecting natural and cultural resources. As a part of the VUM framework, PA managers should strive to understand visitor experiences and attempt to determine appropriate management responses to both prevailing and predicted conditions.

The VUM framework contains critical processes designed to allow PA managers to make more informed and defensible decisions. First, the framework emphasizes understanding who PA visitors are. For instance, a better understanding of visitor motivations would help PA managers match recreation opportunities with recreation needs. Also, knowledge about visitors’ travel behavior is essential to understanding what types of visitor uses are occurring where. Some of the most basic but vital data on outdoor recreation consists of the places people visit, their travel routes, and the amount of time spent at each location (Hallo et al., 2012). Visitors’ movement patterns affect infrastructure and transportation development, the design and maintenance of facilities and services and destination planning (Hallo & Manning, 2010; McGehee et al., 2013). In

order to balance various societal demands and protect natural resources, tracking visitor movement patterns is necessary (Beeco & Brown, 2013). For example, areas where the most use occurs require more intense management, including facility development and redistributing use (Hammitt, Cole & Monz, 2015).

Second, the VUM framework can help determine acceptable levels of impact from visitor use. PA managers need to understand the point(s) where conditions in a recreation area are perceived to be undesirable or degraded. The two critical steps in determining this point(s) include: 1) the identification of indicators (i.e., measurable and manageable variables such as the number of boat encounters per hour) to help define the quality of desired natural/cultural resource conditions and the visitor experience, and 2) defining the acceptable condition of indicator variables (i.e., thresholds). Social science studies to support planning with the VUM framework, or its predecessor frameworks like the Limits of Acceptable Change (LAC) and Visitor Experience Resource Protections (VERP), have most often applied a norm-based approach to model the impact of visitor use on the quality of outdoor experiences (Kuentzel & Heberlein, 2003; Manning, 2011;

Needham, Vaske, Whittaker & Donnelly, 2014). Norms are defined as expectations that individuals and groups use for evaluating ecological and social conditions, and these norms can be useful as a means of formulating indicators and thresholds (Shelby &

Vaske 1991).

Because visitor use fluctuates, but is most often increasing, understanding how a range of visitation levels affects both resources and experiences is helpful for current and future planning. Computer simulations are dynamic and adaptable representation of real

situations that often include consideration of time and/or space (Lawson, Hallo &

Manning, 2008). For example, visitor data collected in 2015 at a visitation level of 100,000 can inform planning efforts for 2020 when visitation is expected to be 150,000 by modelling estimated effects at this visitation level.

Simulation modelling studies that assess social carrying capacity in PAs have to a large extent focused on trails (D’Antonio et al., 2010), attraction sites (Birenboim,

Reinau, Shoval, & Harder, 2015; Bolshakov & Merkuryeva, 2016), and scenic roads (Hallo & Manning, 2010). However, there is limited research that has used simulation modelling in the context of RB (Lowry, Laninga, Zimmerman & Kingsbury, 2011).

Most boat-related simulation studies have examined commercial boating traffic (GeoDimensions, 2006, 2011; Verstichel & Berghe, 2016). Other studies have simulated the capacity of an urban waterway (Itami, 2008) and the potential for vessel collision with large marine wildlife such as whales (Conn & Silber, 2013). Simulation modelling has been used as a tool to evaluate economic and technical issues, risk and accident probabilities, and to perform scenario and policy analyses such as the impact of deepening a river channel (Almaz & Altiok, 2012).

The potential application of simulation modelling to address visitor use issues in the context of RB has been relatively under explored. Yet, simulation research in RB is capable of providing managers with detailed information on both the current and projected boating traffic volumes and densities on specific waterways, and related boating capacities (Itami et al., 2017). By incorporating GPS tracking to collect both motorized and non-motorized boat travel route data, managers can identify “hot spots”

(Lawson, Itami, Gimblett, & Manning, 2006; Beeco & Hallo, 2010). Simulation

modelling may use such data to identify areas capable of accommodating additional use, and also to evaluate risk of possible boating-related accidents or incidences at specific areas.

This paper describes the outcome of a study that applied simulation modelling and the core elements of the VUM framework to the determination of social carrying

capacities on the Middle Delaware Scenic and Recreational River within the Delaware Water Gap National Recreation Area (DEWA). The purpose of this study was guided by the following research questions:

1. What are the indicators and thresholds for quality experiences for RB in DEWA?

2. How do current perceived and observed RB use levels compare to the determined thresholds?

3. Where and at what point would crowding-related thresholds for RB be violated?

4. Are there differences in the model outcomes for perceived crowding thresholds for boating applied on uneven viewsheds?

Literature review The Wild and Scenic Rivers Act (the Act)

The National Wild and Scenic Rivers System (National System) was created by Congress in 1968 to preserve certain rivers or segments of rivers with Outstandingly Remarkable Values (ORVs) — scenic, recreation, geologic, fish and wildlife, historic,

cultural, or other similar values — in a free-flowing condition for the enjoyment of present and future generations (www.rivers.gov). To maintain free flowing status, the construction of dams is prohibited.

Eligible rivers or segments are designated either through an act of Congress or through approval from the Secretary of the Interior if a state governor requests

designation. The latter option requires the river to be first designated as wild, scenic, or recreational (or the equivalent thereof) at the state level by, or pursuant to, an act of the legislature of that state (www.rivers.gov).

Each river is administered by either a federal or state agency. Section 2 (b) of the Act defines the criteria for classification according to the level of development at the time of designation of the shoreline, channel and access as wild, scenic, and/ or recreational (Interagency Wild and Scenic Rivers Coordinating Council [IWSRCC], 2017).

(1) Wild river areas — Those rivers or sections of rivers that are free of impoundments and generally inaccessible except by trail, with watersheds or shorelines essentially primitive and waters unpolluted. These represent vestiges of primitive America.

(2) Scenic river areas — Those rivers or sections of rivers that are free of impoundments, with shorelines or watersheds still largely primitive and shorelines largely undeveloped, but accessible in places by roads.

(3) Recreational river areas — Those rivers or sections of rivers that are readily accessible by road or railroad, that may have some development along their

shorelines, and that may have undergone some impoundment or diversion in the past.

The Act protects the integrity of these rivers while also recognizing the potential for their use and development by promoting cross boundary river management. For example, sections of the Delaware River (i.e., Upper, Middle and Lower) are included in the National System and span across three states, namely, New York (NY), Pennsylvania (PA) and New Jersey (NJ). A wild and scenic river designation

“seeks to protect and enhance a river’s current natural condition and provide for public use consistent with retaining those values. Designation affords certain legal protection from adverse development, e.g., no new dams may be constructed, nor federally assisted water resource development projects allowed that are judged to have an adverse effect on designated river values” (IWSRCC, 2017, p.14).

As of January 2017, some 12,734 miles of 208 rivers have been afforded protection in the National System (IWSRCC, 2017). The Act mandates managing agencies to protect and enhance the ORVs along designated rivers. Therefore, managing agencies must conduct baseline studies, either as part of the process of studying a river for inclusion in the National System or as part of the river management plan drawn up for the river after designation (McGrath, 2014). This has spurred a need to consider the visitor experience, and its management. For example, with 40 miles of the Middle Delaware Scenic and Recreational River within DEWA, boating use studies are essential to guide decision-making.

Visitor management frameworks

The world is biophysically finite, so there is need to have sustainable strategies to limit its use (Hardin, 1968). Measuring carrying capacity is one such strategy. For the past 50 years, outdoor recreation research has adapted and developed the concept of carrying capacity to tackle concerns related to visitor use. Early studies focused on evaluating the number of visitors a recreation area could accommodate before its natural qualities were significantly compromised (Whittaker, Shelby, Manning, Cole & Haas, 2011). Subsequent definitions of capacity introduced the social (i.e., experiential) component by focusing on the quality of the recreation experience (Manning, Lime &

Hof, 1996). An expansion of the capacity concept applied to outdoor recreation led to a three-dimensional approach, comprised of three components: resource, social and management components. These components are interrelated and affect the quality of recreation experiences (Manning, 2011). Incorporating a physical or facility component (i.e., restrictions imposed by limits of physical space) may give a more holistic

representation in determining the capacity of a recreational area (Kim, Shelby &

Needham, 2014). However, the physical component is less often emphasized in the management of outdoor recreation (Manning, 2011; Needham, Ceurvorst, & Tynon, 2013).

Outdoor recreation management frameworks developed to support PAs

management of capacity-related issues include the LAC (Stankey, Cole, Lucas, Petersen

& Frissell, 1985), the Carrying Capacity Assessment Process (CCAP) (Shelby &

Heberlein, 1986), Visitor Impact Management (VIM) (Graefe, Kuss, & Vaske, 1990),

VERP (NPS, 1997) and the VUM framework (IVUMC, 2016). These frameworks employ a management-by-objectives approach to identify and develop indicators and thresholds of quality experiences. They also track the effects of management practices or actions (Manning 2011; Manning, Rovelstad, Moore, Hallo, & Smith, 2015).

The more recent VUM framework, incorporates lessons learned from agency experience (including legal challenges) to allow flexibility in the planning process (Marion, 2016). The framework is divided into four major elements: (1) build the foundation; (2) define visitor use management direction; (3) identify management strategies; and (4) implement, monitor, evaluate and adjust. Regardless of the managing agency, these basic elements are applicable across many visitor use management plans.

Included in each element are steps that provide more detailed direction on the various management topics. For example, when ‘defining visitor use management direction’ (i.e., element 2), PA managers need to 1) define desired conditions for the PA, 2) define appropriate visitor activities, facilities and services, and 3) select indicators and establish thresholds.

The core elements of the VUM framework have been used to guide the

development of DEWA’s Visitor Use Management Plan (VUMP). Specifically, this study addresses the framework’s process of selecting indicators and establishing thresholds in DEWA for the Middle Delaware Scenic and Recreational River.

Normative theory applied to outdoor recreation

In defining the quality of the recreation experience, it is necessary to determine the variables important to visitors. Researchers have given attention to the identification

of these potential indicators of quality for a variety of use types and recreation settings (Manning, 2011). For recreational boating potential indicators may include: the number of boats at one time (BAOT) at attraction sites, number of boats encountered per day, litter, and noise levels (Manning, 2011).

The threshold associated with a particular indicator may simply be developed using logic and professional judgement to estimate what level of change in conditions would prompt more management attention and investment (Cole & Carlson, 2010).

However, where the potential for controversy and the consequences of capacity decisions are high, it might be necessary to use empirical methods.

By using quantitative methods like visitor surveys (Hallo, Brownlee, Hughes, Fefer, & Manning, 2018; Manning et al., 2015), qualitative methods such as conducting interviews (Glaspell, Watson, Kneeshaw, & Pendergrast, 2003) or mixed method

approaches (Hallo & Manning, 2009), managers can obtain data about level and type of use, indicators, users’ personal thresholds and also estimate acceptable conditions in a recreation setting (Manning, 2011).

Studies have often applied a norm-based approach to model the impact of visitor use on the quality of the outdoor experience (Kuentzel and Heberlein, 2003; Manning, 2011; Needham et al., 2014). Normative theory and methods developed in sociology have guided research on evaluating recreational thresholds (Manning, 2013; Pierce &

Manning, 2015). Norms are defined as thresholds that individuals and groups use for evaluating ecological and social conditions (Shelby & Vaske 1991). Using Jackson’s (1965) Return Potential Model (RPM), personal norms of individuals (e.g., boaters) are

aggregated to derive social norms (Manning, 2013) that are often presented graphically in the form of social norm curves as illustrated in Figure 3.1.

Crystallization is a concept for measuring the degree of consensus about

recreation-related norms or the amount of dispersion around a social norm (Krymkowski, Manning, & Valliere, 2009). The vertical line marked by bars at the ends in Figure 3.1 represents crystallization at a single measured point along the norm curve. Standard deviation and variance are some of the most commonly used measures of

dispersion. However, because these measures can make a skewed distribution appear similar to one where there is a uniform agreement and variables can be measured on different scales, concepts like standard deviation are often not comparable across different studies (Krymkowski et al., 2009). The Potential for Conflict Index (PCI2) not only addresses the above issues, it also simultaneously describes a variable's central tendency, dispersion and shape using a graphic display (Manfredo, Vaske & Teel, 2003).

(INSERT FIGURE 3.1 HERE)

PCI2 scores range from 0 to 1. Scores closer to zero indicate higher levels of crystallization or more agreement. The lowest amount of crystallization (agreement) occurs when responses are equally divided between two extreme values such as where 50% of the responses are highly unacceptable and 50% highly acceptable (Engel, Vaske, Bath, & Marchini, 2017). However, there is complete agreement or the highest level of crystallization if all responses (100%) were at any one point on the scale (Engel et al.,

2017). Studies have often used PCI2 for reporting crystallization (Marin, Newman, Manning, Vaske & Stack, 2011; Miller & Freimund, 2017). Also, programs for calculating and graphing PCI2 values are freely available from

http://welcome.warnercnr.colostate.edu/.

A normative approach and related empirical methods have increasingly been used to formulate thresholds in PAs such as DEWA (Brownlee, Sharp, Peterson, & Cribbs, 2018; Hallo, Fefer & Riungu, 2017). Narrative/numerical description of ecological or experiential conditions have been used to measure norms. For example, respondents are asked to evaluate the acceptability of alternative use levels, such as a number of groups encountered per day. The resulting data are then aggregated to determine social norms.

However, narrative/numerical approaches often require respondents to imagine the resource and experiential conditions after reading long descriptive narratives.

Visual-based methods are also commonly used for investigating normative evaluations of recreation settings (Gibson et al., 2014). Studies have used computer generated photo simulations (Gibson et al., 2014), moving images (Kim & Shelby, 2009), video (Freimund, Vaske, Donnelly & Miller, 2002), and images and sound (Grau &

Freimund, 2007) to simulate resource and social conditions for evaluation. These methods are robust and can easily and effectively capture variables that would be awkward to describe using narrative methods (Manning & Freimund, 2004).

Visual-based methods have been used for more than 25 years (Shelby & Harris, 1985) and studies have often supported the validity of these methods to evaluate quality standards at recreation sites (Manning, 2011). Therefore, this paper used computer

generated photo simulations to determine the thresholds for quality boating experiences at DEWA’s portion of the Delaware River.

generated photo simulations to determine the thresholds for quality boating experiences at DEWA’s portion of the Delaware River.